Re: Dynamic adjustments of build dependencies for the number of available processors

[Tim Murphy]

On 5 January 2015 at 18:13, SF Markus Elfring <address@hidden> wrote:

> I assume that you want something different from this
> but you need to explain better, sorry :-)

I hope that an other wording will be clearer.

Can make rules be extended on demand while a build script
is evaluated?

How much can a command like "getconf _NPROCESSORS_ONLN"
influence rule evaluation directly during a make run?

Ok, well I'll make a guess at what you want:

    You have some sort of work which you can control the "granularity" i.e. you can either have 1 rule that generates 10 targets or you can have 10 rules that generate 1 target.

Normally I'd just say "have 10 rules that generate 10 targets" because this is the simplest thing to do and works properly within make and is what people expect. But I can imagine ways in which this would not be efficient - particularly if the startup time for the rule was high.

So you have in your toolbox $(shell) and $(eval).

$(eval) lets you generate rules dynamically.

$(shell) lets you run an external tool in a shell

e.g. you could use $(shell) to find out how many processors there were like this (linux):

NPROC:=$(shell grep "processor" /proc/cpuinfo  | cut -b12- )

# this returns "0 1 2 3" etc

# Then you might use eval to generate rules dynamically:

$(foreach N,$(NPROC),$(eval

   # YOUR DYNAMIC RULE

  MYTARGET_$(N):

<TAB>mytool --dosomething -o$@

  )

)

I am probably guessing wrong but perhaps this gives you some ideas.

Regards,

Tim

--

You could help some brave and decent people to have access to uncensored news by making a donation at:

http://www.thezimbabwean.co.uk/friends/